This work investigates the role of the temporal dimension in the processes underlying behavior as it unfolds in real time. A theoretical schema has been proposed in which the central notion is that the temporal organization of a behavior sequence is relative timing, (rhythmic patterning), as opposed to concatenation, of elements in the sequence. Among the specific objectives of this work are the following (a) to test a proposed speaker-listener model against implications from alternative models, including analysis-by-synthesis and filter models, by means of experiments in which the input signal is temporally distorted in various ways, (b) to apply temporal-patterning notions to several practical problems, including speech synthesis and teaching of reading to language- handicapped children.